On 12/17/2011 11:47 AM, Eric Smith wrote:
Mouse wrote:
You mentioned OpenGenera; is this an open-source
version or something?
[...]
My (rudimentary) web-search skillz end up at
pages that give the
impression the "Open" part is a misnomer,
Liam Proven wrote:
Kinda, yes. It certainly doesn't mean what it
normally tends to.
Like OpenVMS.
Shortly after the name change to OpenVMS was announced, I received in
the mail a brochure for training classes DEC offered. One of the
classes was on the subject of writing OpenVMS device drivers. It was
noted that a signed nondisclosure agreement was required to take the
class. That's an even less "open" operating system than Microsoft
Windows.
If you took the course you got to see DEC developed drivers in the
flesh. Hence NDA. OVMS was propritory but the interfaces to the OS
matched the open concept so developers could work with it without NDA.
M$ did and still does have interfaces that are both undocumented
they also use backdoor interfaces, never minding the interfaces that
were poorly documented or didn't work as described or would change
with updates.
DEC marketing was usurping the word "Open"
because they noticed that
customers were starting to want open things. They
apparently thought
(probably correctly) that putting "Open" in the name would trick
potential customers and bring in more sales. Other companies did the
same thing, but I can't think of any others where the oxymoronic name
has lasted as long.
One may feel that way but during a time of OSs that were so locked up
you needed NDA and developers kit from that vendor before you
could start it was the right direction.
Someone suggested that in this usage, "Open" isn't an adjective, but
rather a verb, and the direct object isn't "VMS" but rather your wallet.
From the hobby perspective Open always meant Free, and Sources. The
professionals it meant access to what was needed and
good documentation and training as the OS code was not their job nor
anything they wanted to mess with unless broken and
for that they expected a level of support that could resolve that.
On the other hand VMS was/is trying to maintain a reputation of 24x365
and M$ was reboot often and keep the reset switch on the
front of the machine. To have a machine that can do 24x365 you need
good hardware and all along there were PCs up to the task
but the OSs common were not so robust. From my perspecive at one time I
had the luck of running VMS, One of the sorta unix (on pC)
servers and NT4.51, VMS the biggest job was keeping the power supplied
to the box and occasion new product additions, The unix PC
was fairly stable but would suffer from disk and IO issues, the NT3.51
box was same hardware and added BSOD and memory leaks
requiring reboots every week or application hangs never minding keeping
at it to insure security. Also the VMS machine I could
develop and deploy often without neededing to write a line of compiled
code as DCL was usually more than enough.
The other side of the argument was DEC licenses VMS and with that
purchase you got a wall of documentation. The level of
documentation runs from user level and system administrator to outright
this is what you need to write an app for VMS with
all the supporting information. M% and even LINUX does not match that.
Granted Linux you can just find the module and
read the code but your subject to the writing style of the author and
the available documentation and there is the little issue
of what was intended vs what it actually did. In the end many
programmer do not have the time to wade through hundreds
or even thousands of lines of code for a OS bug or misfeature.
In the end there was a unix core OS for PCs that was good but who
remembers them and what ever happened to them.
hint the protocol for networking was IPX because they didn't do IP or
DECnet natively. Then there was OS2, SCO Unix
and so on. The only PC (WINTEL) based OS that has made inroads on the
WINTEL world is Linux and I'd bet 85% of the
users cannot compile anything on it, nor wish too, as they use it as
applications platform.
Allison